通过Messenger Distillation实现个性化医疗保健分析的异构协作学习
Heterogeneous Collaborative Learning for Personalized Healthcare Analytics via Messenger Distillation
摘要:
在本文中,我们提出了一个基于相似性质量的Messenger Distillation(SQMD)框架,用于异构异步设备上的医疗保健分析。通过引入预加载的参考数据集,SQMD使所有参与设备能够通过信使(即,由客户端生成的参考数据集的软标签),而不假设相同的模型架构。此外,信使还携带重要的辅助信息来计算客户端之间的相似性和评估每个客户端模型的质量,基于此,中心服务器创建并维护动态协作图(通信图),以提高异步条件下SQMD的个性化和可靠性。在三个真实数据集上的大量实验表明,SQMD取得了优异的性能。
英文摘要:
In this paper, we propose a Similarity-Quality-based Messenger Distillation (SQMD) framework for heterogeneous asynchronous on-device healthcare analytics. By introducing a preloaded reference dataset, SQMD enables all participant devices to distill knowledge from peers via messengers (i.e., the soft labels of the reference dataset generated by clients) without assuming the same model architecture. Furthermore, the messengers also carry important auxiliary information to calculate the similarity between clients and evaluate the quality of each client model, based on which the central server creates and maintains a dynamic collaboration graph (communication graph) to improve the personalization and reliability of SQMD under asynchronous conditions. Extensive experiments on three real-life datasets show that SQMD achieves superior performance.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。